Parse each word’s morphosyntactic function, grammatical gender, number, lemma, etc.
Identify a document’s relevant domains through a specialized terminology lexicon.
Annotate texts using lexicon’s terms that are directly related to a given field.
Isolate expressions designating named entities (people, places, organizations).
Train a prediction model to note the properties specific to a type of annotation by annotating occurrences that provide a maximum of information for the model, allowing to annotate with minimal effort.
CRIM receives funding under the CANARIE Research Software Program to develop the PACTE platform.
Alpha version available on line, three research groups participating.
Presentation of PACTE platform and domain profiling annotation service at the 12th International Conference on Computational Semantics (IWCS 2017) for the "Interoperable Semantic Annotation" and the "Language, Ontology, Terminology and Knowledge Structures" workshops (Montpellier, France).
Presentation of the named entity annotation service at the Text Analysis Conference 2017 in Maryland.
Be part of our story!
Researcher - Natural Language Processing
Speech and Text Team
Senior Researcher – Natural Language Processing
Speech and Text Team
Director
Research and Development
Postdoctoral Researcher – Natural Language Processing
Speech and Text Team
Advisor
Speech and Text Team
Senior research agent
Speech and Text Team
Senior Research Agent
Emerging Technologies and Data Science Team
Senior Research Agent
Emerging Technologies and Data Science Team
Senior Research Agent
Emerging Technologies and Data Science Team
Research Agent
Emerging Technologies and Data Science Team
Research Agent
Emerging Technologies and Data Science Team
Research Agent
Emerging Technologies and Data Science Team
Front-End Developer
Vision and Imaging Team